Loading…

Structured Forests for Fast Edge Detection

Edge detection is a critical component of many vision systems, including object detectors and image segmentation algorithms. Patches of edges exhibit well-known forms of local structure, such as straight lines or T-junctions. In this paper we take advantage of the structure present in local image pa...

Full description

Saved in:

Bibliographic Details
Main Authors:	Dollar, Piotr, Zitnick, C. Lawrence
Format:	Conference Proceeding
Language:	English
Subjects:	Algorithms Computer vision Decision trees Detectors Edge detection Image color analysis Image edge detection Image segmentation Learning realtime vision Segmentation State of the art structure learning Training Vegetation
Citations:	Items that cite this one
Online Access:	Request full text
Tags:	Add Tag No Tags, Be the first to tag this record!

cited_by	cdi_FETCH-LOGICAL-c296t-fe6c3f15847e7c9f14349ebbeab58a793af4bddac5578348bc8ab23129cfe41d3
cites
container_end_page	1848
container_issue
container_start_page	1841
container_title
container_volume
creator	Dollar, Piotr Zitnick, C. Lawrence
description	Edge detection is a critical component of many vision systems, including object detectors and image segmentation algorithms. Patches of edges exhibit well-known forms of local structure, such as straight lines or T-junctions. In this paper we take advantage of the structure present in local image patches to learn both an accurate and computationally efficient edge detector. We formulate the problem of predicting local edge masks in a structured learning framework applied to random decision forests. Our novel approach to learning decision trees robustly maps the structured labels to a discrete space on which standard information gain measures may be evaluated. The result is an approach that obtains real time performance that is orders of magnitude faster than many competing state-of-the-art approaches, while also achieving state-of-the-art edge detection results on the BSDS500 Segmentation dataset and NYU Depth dataset. Finally, we show the potential of our approach as a general purpose edge detector by showing our learned edge models generalize well across datasets.
doi_str_mv	10.1109/ICCV.2013.231
format	conference_proceeding
fullrecord	<record><control><sourceid>proquest_CHZPO</sourceid><recordid>TN_cdi_proquest_miscellaneous_1669889217</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>6751339</ieee_id><sourcerecordid>1669889217</sourcerecordid><originalsourceid>FETCH-LOGICAL-c296t-fe6c3f15847e7c9f14349ebbeab58a793af4bddac5578348bc8ab23129cfe41d3</originalsourceid><addsrcrecordid>eNotzj1PwzAUhWGDQKItjEwsGRFSiq8_Yt8RhRYqVWLgY40c5xoFtQ3YzsC_J1KZzvLo6GXsGvgSgOP9pq4_loKDXAoJJ2wOyiAKq7g4ZTMhLS-N5uqMzUBrXmqFeMHmKX1xLidWzdjda46jz2OkrlgPkVJORRhisXYpF6vuk4pHyuRzPxwu2Xlwu0RX_7tg7-vVW_1cbl-eNvXDtvQCq1wGqrwMoK0yZDwGUFIhtS25VltnULqg2q5zXmtjpbKtt66d6gX6QAo6uWC3x9_vOPyMU1Gz75On3c4daBhTA1WF1qIAM9GbI-2JqPmO_d7F36YyGqRE-Qe_yFAq</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype><pqid>1669889217</pqid></control><display><type>conference_proceeding</type><title>Structured Forests for Fast Edge Detection</title><source>IEEE Xplore All Conference Series</source><creator>Dollar, Piotr ; Zitnick, C. Lawrence</creator><creatorcontrib>Dollar, Piotr ; Zitnick, C. Lawrence</creatorcontrib><description>Edge detection is a critical component of many vision systems, including object detectors and image segmentation algorithms. Patches of edges exhibit well-known forms of local structure, such as straight lines or T-junctions. In this paper we take advantage of the structure present in local image patches to learn both an accurate and computationally efficient edge detector. We formulate the problem of predicting local edge masks in a structured learning framework applied to random decision forests. Our novel approach to learning decision trees robustly maps the structured labels to a discrete space on which standard information gain measures may be evaluated. The result is an approach that obtains real time performance that is orders of magnitude faster than many competing state-of-the-art approaches, while also achieving state-of-the-art edge detection results on the BSDS500 Segmentation dataset and NYU Depth dataset. Finally, we show the potential of our approach as a general purpose edge detector by showing our learned edge models generalize well across datasets.</description><identifier>ISSN: 1550-5499</identifier><identifier>EISSN: 2380-7504</identifier><identifier>EISBN: 1479928402</identifier><identifier>EISBN: 9781479928408</identifier><identifier>DOI: 10.1109/ICCV.2013.231</identifier><identifier>CODEN: IEEPAD</identifier><language>eng</language><publisher>IEEE</publisher><subject>Algorithms ; Computer vision ; Decision trees ; Detectors ; Edge detection ; Image color analysis ; Image edge detection ; Image segmentation ; Learning ; realtime vision ; Segmentation ; State of the art ; structure learning ; Training ; Vegetation</subject><ispartof>2013 IEEE International Conference on Computer Vision, 2013, p.1841-1848</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c296t-fe6c3f15847e7c9f14349ebbeab58a793af4bddac5578348bc8ab23129cfe41d3</citedby></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/6751339$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>310,311,315,786,790,795,796,2071,27957,27958,54906,55271,55283</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/6751339$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Dollar, Piotr</creatorcontrib><creatorcontrib>Zitnick, C. Lawrence</creatorcontrib><title>Structured Forests for Fast Edge Detection</title><title>2013 IEEE International Conference on Computer Vision</title><addtitle>iccv</addtitle><description>Edge detection is a critical component of many vision systems, including object detectors and image segmentation algorithms. Patches of edges exhibit well-known forms of local structure, such as straight lines or T-junctions. In this paper we take advantage of the structure present in local image patches to learn both an accurate and computationally efficient edge detector. We formulate the problem of predicting local edge masks in a structured learning framework applied to random decision forests. Our novel approach to learning decision trees robustly maps the structured labels to a discrete space on which standard information gain measures may be evaluated. The result is an approach that obtains real time performance that is orders of magnitude faster than many competing state-of-the-art approaches, while also achieving state-of-the-art edge detection results on the BSDS500 Segmentation dataset and NYU Depth dataset. Finally, we show the potential of our approach as a general purpose edge detector by showing our learned edge models generalize well across datasets.</description><subject>Algorithms</subject><subject>Computer vision</subject><subject>Decision trees</subject><subject>Detectors</subject><subject>Edge detection</subject><subject>Image color analysis</subject><subject>Image edge detection</subject><subject>Image segmentation</subject><subject>Learning</subject><subject>realtime vision</subject><subject>Segmentation</subject><subject>State of the art</subject><subject>structure learning</subject><subject>Training</subject><subject>Vegetation</subject><issn>1550-5499</issn><issn>2380-7504</issn><isbn>1479928402</isbn><isbn>9781479928408</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2013</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNotzj1PwzAUhWGDQKItjEwsGRFSiq8_Yt8RhRYqVWLgY40c5xoFtQ3YzsC_J1KZzvLo6GXsGvgSgOP9pq4_loKDXAoJJ2wOyiAKq7g4ZTMhLS-N5uqMzUBrXmqFeMHmKX1xLidWzdjda46jz2OkrlgPkVJORRhisXYpF6vuk4pHyuRzPxwu2Xlwu0RX_7tg7-vVW_1cbl-eNvXDtvQCq1wGqrwMoK0yZDwGUFIhtS25VltnULqg2q5zXmtjpbKtt66d6gX6QAo6uWC3x9_vOPyMU1Gz75On3c4daBhTA1WF1qIAM9GbI-2JqPmO_d7F36YyGqRE-Qe_yFAq</recordid><startdate>20131201</startdate><enddate>20131201</enddate><creator>Dollar, Piotr</creator><creator>Zitnick, C. Lawrence</creator><general>IEEE</general><scope>6IE</scope><scope>6IH</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIO</scope><scope>7SC</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope></search><sort><creationdate>20131201</creationdate><title>Structured Forests for Fast Edge Detection</title><author>Dollar, Piotr ; Zitnick, C. Lawrence</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c296t-fe6c3f15847e7c9f14349ebbeab58a793af4bddac5578348bc8ab23129cfe41d3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2013</creationdate><topic>Algorithms</topic><topic>Computer vision</topic><topic>Decision trees</topic><topic>Detectors</topic><topic>Edge detection</topic><topic>Image color analysis</topic><topic>Image edge detection</topic><topic>Image segmentation</topic><topic>Learning</topic><topic>realtime vision</topic><topic>Segmentation</topic><topic>State of the art</topic><topic>structure learning</topic><topic>Training</topic><topic>Vegetation</topic><toplevel>online_resources</toplevel><creatorcontrib>Dollar, Piotr</creatorcontrib><creatorcontrib>Zitnick, C. Lawrence</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan (POP) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE/IET Electronic Library</collection><collection>IEEE Proceedings Order Plans (POP) 1998-present</collection><collection>Computer and Information Systems Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Dollar, Piotr</au><au>Zitnick, C. Lawrence</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Structured Forests for Fast Edge Detection</atitle><btitle>2013 IEEE International Conference on Computer Vision</btitle><stitle>iccv</stitle><date>2013-12-01</date><risdate>2013</risdate><spage>1841</spage><epage>1848</epage><pages>1841-1848</pages><issn>1550-5499</issn><eissn>2380-7504</eissn><eisbn>1479928402</eisbn><eisbn>9781479928408</eisbn><coden>IEEPAD</coden><notes>ObjectType-Article-2</notes><notes>SourceType-Scholarly Journals-1</notes><notes>ObjectType-Conference-1</notes><notes>ObjectType-Feature-3</notes><notes>content type line 23</notes><notes>SourceType-Conference Papers & Proceedings-2</notes><abstract>Edge detection is a critical component of many vision systems, including object detectors and image segmentation algorithms. Patches of edges exhibit well-known forms of local structure, such as straight lines or T-junctions. In this paper we take advantage of the structure present in local image patches to learn both an accurate and computationally efficient edge detector. We formulate the problem of predicting local edge masks in a structured learning framework applied to random decision forests. Our novel approach to learning decision trees robustly maps the structured labels to a discrete space on which standard information gain measures may be evaluated. The result is an approach that obtains real time performance that is orders of magnitude faster than many competing state-of-the-art approaches, while also achieving state-of-the-art edge detection results on the BSDS500 Segmentation dataset and NYU Depth dataset. Finally, we show the potential of our approach as a general purpose edge detector by showing our learned edge models generalize well across datasets.</abstract><pub>IEEE</pub><doi>10.1109/ICCV.2013.231</doi><tpages>8</tpages></addata></record>
fulltext	fulltext_linktorsrc
identifier	ISSN: 1550-5499
ispartof	2013 IEEE International Conference on Computer Vision, 2013, p.1841-1848
issn	1550-5499 2380-7504
language	eng
recordid	cdi_proquest_miscellaneous_1669889217
source	IEEE Xplore All Conference Series
subjects	Algorithms Computer vision Decision trees Detectors Edge detection Image color analysis Image edge detection Image segmentation Learning realtime vision Segmentation State of the art structure learning Training Vegetation
title	Structured Forests for Fast Edge Detection
url	http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-09-22T21%3A35%3A26IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_CHZPO&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Structured%20Forests%20for%20Fast%20Edge%20Detection&rft.btitle=2013%20IEEE%20International%20Conference%20on%20Computer%20Vision&rft.au=Dollar,%20Piotr&rft.date=2013-12-01&rft.spage=1841&rft.epage=1848&rft.pages=1841-1848&rft.issn=1550-5499&rft.eissn=2380-7504&rft.coden=IEEPAD&rft_id=info:doi/10.1109/ICCV.2013.231&rft.eisbn=1479928402&rft.eisbn_list=9781479928408&rft_dat=%3Cproquest_CHZPO%3E1669889217%3C/proquest_CHZPO%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c296t-fe6c3f15847e7c9f14349ebbeab58a793af4bddac5578348bc8ab23129cfe41d3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=1669889217&rft_id=info:pmid/&rft_ieee_id=6751339&rfr_iscdi=true